POSTOPERATIVE ATELECTASIS AND BRONCHOPNEUMONIA

OBSERVATIONS ON THE PREVENTION AND TREATMENT OFPOSTOPERATIVE ATELECTASIS AND BRONCHOPNEUMONIA*

CAMERON HAIGHT, M.D.,AND

HENRY K. RANSOM, M.D.ANN ARBOR, MICH.

FRONM TIlE D)EPARTMENT OF SURGERY, UNIVERSITY OF MICHIGAN, ANN ARBOR, MIICII.

THE POSTOPERATIVE pulmoniary comiiplications that are usually enicounteredare of two general types: those due to the retention of bronclhial secretionsan(d those due to emboli. Althouglh emboli8' 9, 10, 12 are responsible for a cer-tain percentage of complicationis, we believe the great majority of instancesof postoperative pnieumonia and atelectasis are the result of retained bronchialsecretions. The emphasis in this paper will, therefore, be upon the complica-tions resulting fromii the retenition of bronchial secretions. Particular con-sideration will be given to the early recognition of the presence of bronchialsecretions, to the measures for preveniting the retention of secretions, andto the methods for promoting bronchial drainage when retained secretionsare present. When bronchial secretions occur as a result of septic or asepticemboli, they must be nmanaged in the same way as secretions due to othercauses.

The development of postoperative complications when due to retainedbronchial secretions is dependent upoIn two causes. One is the presence ofbronchial secretions, which may be due to a mild preexisting inflammation ofthe respiratory tract, to the entrance of pharyngeal secretions into the bron-chial tree either during or following operation, or to the formation of bronchialsecretions postoperatively. The second cause is the decreased respiratory andcough efficiency associated with the operation.

The incidence of postoperative atelectasis and bronchopneumonia is im-portantly influenced by several factors. The location of the operative sitehas an important influence in determining the frequency with which thesecomplications are encountered. They are noted most often following upperabdominal operations,20 and their incidence following gastric operations isabout twice as great as following operations on the biliary system. Also, theyoccur more frequently in patients who have chronic cough25 and expectorationbefore operation, as symptoms of chronic bronchitis, bronchiectasis, para-nasal sinusitis and asthma, as well as in patients with acute respiratory infec-tions. Another factor is the sex difference-postoperative atelectasis andbronchopneumonia being two or two and one-half times as frequent in malesas in females. This difference has been attributed to the fact tllat, normally,females are predominately costal breathers, wlhereas males are predominately

* Read before the American Surgical Association, White Sulphur Springs, W. Va.,April 30, I941.

243

HAIGHT AND RANSOM Annals of Surge14

diaphragmatic and abdominal breathers. As a result, the postoperative dia-phragmatic splinting exerts a more profound reduction in the respiratoryefficiency of males than it does in females. Beecher2 has shown that thereduction in vital capacity following upper abdominal operations is greaterin men (58 per cent) than in women (49 per cent), and that the greatestand most rapid rate of recovery takes place between the second and fourthday in women, while in males it is delayed and takes place between the fourthand sixth days. These observations suggest that the greater incidence in malesis due to the greater reduction in pulmonary ventilation and to the greaterredluction in the efficiency of respiration and coughing after operation.Another possible explanation for the greater incidence in males is the morefrequent incidence of cough and expectoration preoperatively.

PREVENTION

As the preoperative condition of the respiratory tract miiay influenlce the(leveloplment of postoperative pulmonary complications, the possibility of asmall amount of daily sputumii should be sought for by careful questioning,' 5since this is often present and not appreciated or acknowledged by the patientunless he is particularly questioned about it. If sputum is present, care shouldbe taken imimiediately before the anesthesia is begun to be certain that allsecretionis have been raised. Furthermore, if a history of expectoration hasbeen elicited, the necessity for maintaining constanit remiioval of these secre-tions after operation is obvious. When an acute respiratory infection ispresent, operations of election should be delayed for a considerable periodfollowing recovery from the respiratory infection. A delay of only a fewdays is usually not sufficient to allow complete recovery, and the operationpreferably should be deferred until at least two or more weeks have elapsedafter the acute symptoms have disappeared. Faulty oral hygiene should becorrected prior to operations of election, even when local or spinal anesthesiais used, for the reason that oral and pharyngeal secretions may gravitate intothe tracheobronclhial tree during sleep, unless the posture of the patient isregulated to prevent this occurrence.

The type of preoperative sedation should be chosen with care, and thosesedatives which cause prolonged sleep or drowsiness after operation shouldbe avoided. A large dose of a long-acting barbiturate, such as nembutal, isinadvisable, and if a barbiturate need be used, it should preferably be of abrief-acting type. The advisability of administrating atropine preoperativelyhas been questioned by many persons because it inicreases the viscosity ofbronchial secretions, and we believe it should be avoided unless indicatedfor a definite reason, such as the prevention of excessive salivation duringoperations in the region of the mloutlh or pharynx.

OperatiVe Conisi(derationis.-Tlhe position of the patielnt oni the operatingtable is of importanice in preveniting the gravity (Irainiage of oral andI plharyln-geal secretions into the tracheobronichlial tree. Especially in patients withpharynigeal secretions, the head should not be raisedl on a pillow as this posi-tion allows a better opportunity for the secretions to enter the trachea. In

244

Voliime 114Numiiibei:2

ATELECTASIS AND BRONCHOPNEUNIONIA

or(ler to oxvercolle tlle irl).al Iosterior deviatioll of thle trachiea, the operatingtal)le s1ioti(1 b)e iiclille(l to a 1o or 1 5leg-re Tlreldelibl)urgopositiolL. Thenormial curve of thle upper d(lorsal spline varies, beiing as a rule miore mial-ke(din el(lerly 1)atienits. The trachea accordingly lhas a greater posterior (leviationin elderly than young persons andl tlherefore, as a rule the greater degree ofTrendelenburg positioni should be used in elderly persons. However, thetype of operation or the condition for whiclh it is dcone may demanld that thetable be flat or the head of the table elevated.

The available evidence suggests that the incidcence of atelectasis and bron-chopneumonia is not importantly influenced by the anestlhetic agent. It isdesirable, however, that the anesthesia should be gauged so that the patientwill awaken promptly following the operation. Schmidt and \Vaters27 oh-served that the greatest incidence of pulmonary complications followedl theuse of ether, and the next greatest incidence followed spinal anesthesia.Brown,4 however, is of the opinion that the incidence of pulmonary atelectasisis greater following spinal anestlhesia than witlh any form of inhalation orregional anesthesia. He attributes this to the fact that spinal anesthesia defi-nitely inhlibits the depth and force of respiratory movements not only duringthe operation but for a considerable period tlhereafter. Brown also believesthat it is these respiratory movemlents (both intrinsic and extrinsic) whichtend to rid the tracheobronclhial tree of foreign matter or secretion.

Jones and McClure,19 in 1931, called attention to the influence of thetransverse upper abdominal incision in reducing the incidence of postoperativecomplications. As the transverse incision is in the plane of the muscular andaponeurotic fibers of the external and internal oblique and transversusabdominiis muscles, the pull of these respiratory muscles during costal excur-sioIn tends to approximate and relax the wound rather than to exert tensionupon it, as witlh the vertical incision. Furthermore, the transverse incisionoffers less opportunity for injury to the intercostal nerves supplying themusculature of the upper abdominal wall. Jones and McClure observed areduction of pain and more nearly normal respiratory excursion and pul-monary ventilation after the use of the transverse incision. They were im-pressed by the comfort and ease of breathing in the average short, obesepatient following a gallbladder operation witlh the transverse incision. In aseries of I25 consecutive transverse abdominal incisions, most of which werefor operations upoIn the gallbladder, bile ducts and stomach, they noted noinstaince of postoperative pneumoniia or atelectasis. Five patients developedpulionary embolismi which proved fatal in two instances. Jones and McClurestate that two or tlhree additional patients developed pulmonary symptoms,such as couglh of mild or moderate degree, hut other evidences definitelyestablishing a pulmonary complicationi were lacking. Their statistics are instriking contrast to the usual incidence of pulmonary atelectasis and broncho-pneumonia, wlhich is 5 to 12 per cent and occasionally more, in most series ofgallbladder operations. During the last few years the transverse incision hasbeen used at the University of Michigan Hospital witlh increasing frequencyin gallbladder anid other abdominal operations. and at present it is used almost

245

HAIGHT AND RANSOM Annals of SurgeryHAIGHTANDRANSOM ~~August, 1941

routinely in gallbladder operations. Our observations in these cases agreewith those of Jones and McClure in regard to the smoother convalescence.In a series of operations for chronic cholecystitis and cholelithiasis, we havefound the incidence of postoperative atelectasis and pneumonia considerablylower than with the vertical incision, but not as low as reported by Jones andMcClure.

Postoperative Considerations.-It is highly desirable that the anesthesiaand the preoperative medication should allow patients to awaken promptlyfollowing operation, thereby enabling them to cough and expectorate by thetime they are to be placed in bed. In the event that the respirations are"wet" and that consciousness is not resumed shortly after operation, or that

.,

'^_ i}S- _~~~~~~~~~~~~~~... .... ..' ..: .. ......~~~~...,:

FIG. I.-Lateral position. Frontal bronchogram demonstrating dependent drainage ofbronchial tree of uppermost lung with patient on contralateral side.

coughing is ineffectual, aspiration of the tracheobronchial tree by cathetersuction or by bronchoscopy is indicated. It is believed that hyperventilationwith carbon dioxide and oxygen at the conclusion of the operation is in itselfnot sufficient if secretions are present and cannot be raised by expectoration.When hyperventilation is used, it is generally agreed by anesthetists that anonabsorbable buffer agent, such as nitrogen, helium or air, should be used inthe inhaled gas mixture so that alveolar collapse will not result from the soleuse of two rapidly absorbable gases such as carbon dioxide and oxygen.

Elevation of the foot of the bed13 is advisable until consciousness isresumed, unless circumstances contraindicating this position are present. Aconsiderable degree of the Trendelenburg position is necessary to secure actualgravitation of tenacious secretions toward the pharynx.' The foot of the bedmay need to be elevated ten to i8 inches (seven to i i degrees) or more tocause the trachea to become horizontal when the patient is in the supine posi-tion. Even so, the posterior segment of the upper lobe and the superiordorsal and subapical segments of the lower lobe remain dependent. For thisreason, it is advisable that the patient be turned on alternate sides26 in orderthat the maximum benefit of posture will be obtained for each lung duringthe time that it is uppermost (Fig. I). As the lower portion of the trachea

246

Volume 114Number 2 ATELECTASIS AND BRONCHOPNEUMONIA

usually deviates slightly to the right, dependent drainage of the trachea isobtained when the patient is lying flat on the left side, but the foot of the bedmust be elevated to provide dependent tracheal drainage when patient is on

A. B.

Nov. 19, 1938NO.2,13

Nov. 23, 1938 Nov. 28, 1938C. D.

FIG. 2.-M. M., NO. 372675: (A) Roentgenogram, Nov. 19, 1938, on first postoperativeday following total gastrectomy shows partial atelectasis of right lung. Tracheobronchial suc-tion on three occasions during a period of five hours. Improvement of voluntary cough, sopatient treated hy alternate lateral positions, one hour on left aide and one-half hour on right,and by encouragement of frequent cough. (B) Roentgenogram, Nov. 2I, 1938, two dayslater. Right lung almost clear. Pneumonitis has developed on left, Cough productive andeffectual. Subsequent treatment by change of position at hourly and half-hourly intervalswith patient kept one hour on right side and one-half hour on left. (C) Roentgenogram,Nov. 23, 1938, two days later. Return of atelectasis of right lower lobe. Left lung nowclear. Position now changed more frequently than before to prevent stasis of secretions inmore dependent lung. (D) Roentgenogram, Nov. 28, 1938, five days later. Clinical improve.ment with only slight residual pneumonitis at right base.

This case illustrates the beneficial effect of the lateral position in improving the drainageof the lung which is uppermost for the greater length of time, and the possibility of drainageof secretions into the dependent lung. More frequent change of position (at least every 20minutes) would have decreased the tendency for involvement of the more dependent lung.

the right side. After consciousness has returned, the use of the lateral posi-tion should be continued by having the patient lie on alternate sides forperiods of not longer than 30 minutes each, or by alternating the lateral

247

HAIGHT AND RANSOM Annals of SurgeryAugust, 1941

positions with the supine position. The obvious disadvantage of the lateralposition is that it does not provide drainage of the dependent bronchial treeand it also allows secretions from the uppermlost lung to gravitate into thedependent lung,1' unless an uncomfortable (legree of elevation of the footof the bed is maintainied. For this reason, patients should not be allowed toremain on one side for a prolonged period. Ordinarily, the maximum tinmethe patient should lie upon one side should not be more than 20 to 30 miniutes,but slhould be governed by the amount of secretions, being shorter wlhenabundant secretions are present (Fig. 2).

FIG. 3.-Hypoventilation of right lung 36 hours following sigmoidcolostomy. Mloderate abdominal distention. Breath sounds absent atright base until ventilation of base was obtained by coughing.

The recognition of varying degrees of pulmonary lhypoventilationi canbe obtained by physical, as well as roentgeniologic. examinlation of the chest.The physical examlination of the bases of the lunigs is facilitated wlhen thelung being examinied is in an elevated position (witlh the patient lying on theopposite side). as this positioni offers the patiellt an opportunity to demon-strate the maxinmtum voltuntary amlounit of ventilation of the lower lobe of theuppermlost sidle. Any (lecrease in the amounit of ventilation is, therefore, ofgreater significance thani when the physical examinationi of thle bases is carriedlout witlh the l)atient lying on hiis back. Ordinarily, following upper abldominaloperations, considerable hypoventilation of botlh bases exists, being greateron the riglht thani on the left24 (Fig. 3). The dlegree of hlypoventilation is influ-enced by a number of factors, amonig theml being the location and type ofoperation, 2 the severity of the operation. the general condition and sex

248

Voluine 114Number 2 ATELECTASIS AND BRONCHOPNEUMONIA

of the patient, the amount of pain and muscle splinting,6 the amount ofabdominal distention3 and the degree of obesity. It is a usual finding thatthe decreased breath sounds at the bases of the lungs following upper abdom-inal operations can be materially augmented during the examination by havingthe patient take several deep breaths, by the deep inspiration that followsvoluntary cough (Fig. 4) or by hyperventilation with carbon dioxide-oxygeninhalations. We have occasionally seen patients in whom the pulmonaryhypoventilation was so marked that breath sounds over the lower dorsalsegments of the lower lobe could not be heard on deep breathing or aftercough, but could be elicited only by auscultation during hyperventilation with15 per cent carbon dioxide and oxygen. It is particularly recommended thatthe character and intensity of breathsounds be elicited during one or all ofthe above-mentioned maneuvers. If theintensity of the breath sounds reaches 2or approximates normal and if rales andrhonchi are absent, one can be reasonablycertain that the bronchi are not obstructedby the secretions and that the decreasedbreath sounds heard at the beginning ofthe examination were due to hypoventila-tion and not to atelectasis.

Roentgenologic examination of thechest following operations on the abdo-men and particularly on the upper abdo-men showing varying degrees of hypo-ventilation, as is evidenced by elevationof the leaves of the diaphragm, decreaseof costal expansion, and decreased aera-tion of the lungs, especially at the bases.Pulmonary hypoventilation is evidencedby a generalized haziness and loss ofaeration of the lung, especially of thatportion immediately above the dia- m L.c

phragm. Hypoventilation will appear more, . . . ~~~~~~~~~FIG.4.-Examination of base of lungsmarked than it actually is if the roent- is preferably undertaken with patient lying

on the side, in alternate lateral positions,genograms should be made in the ex- thereby improving ventilation of uppermost

lung. Breath sounds that may be decreasedpiratory phase of respiration, an occur- or absent due to hypoventilation, can be aug-mented when auscultation is done during

rence which is noted when patients are cough (illustrated), or during hyperventila-tion with carbon dioxide-oxygen inhalations,unable to cooperate by holding a deep thus aiding in differentiation between atelec-tasis and hypoventilation. Rhonchi due toinspiration while the exposure is being bronchial secretions may be detected by thesemaneuvers when otherwise not audible. Inmade. Therefore, the position of the the illustration the incision for a gastric oper-ation is being supported by the examiner'sleaves of the diaphragm and the amount hand. The tube in the patient's nose is for

of expansion of the thoracic wall should continuous duodenal suction.

be ascertained when interpreting the nature of basal densities. Also, theroentgenologic technic used for exposure of the films must be taken into

249

HAIGHTAND RANSOM ~~~~~~Annals of SurgeryHAIGHT AND RANSOM August. 1941

consideration in interpreting the degree of hypoventilation. As the roent-genograms will ordinarily have been made with a portable unit, and withthe patient in a semireclining positioin, the leaves of the diaphragm will appearhigher and the degree of ventilation will appear less than if they had beenmade with the patient in an upriglht position and with a standard chest unit.If considerable hypoventilation is present, the increased density of the basesof the lungs may simulate patchy atelectasis or bronchopneumonia.24

Hypoventilation decreases the effectiveness of bronchial drainage by anactual reduction in the size and motility of the bronchi and by a reductionin the amount of air that can be expelled fromii the lungs by coughing, Thereduced diameter of the bronclhi interferes witlh the drainage of viscid bron-clhial secretions and lessens the to-and-fro movemiient of air which in itselfaids drainage. The hypoventilation of the lungs lessens the available amountof air that can be expelled by couglhing, tlhereby reducing the effectivenessof expectoration by decreasing the volume and force of the coughing act.MIeasures that aid in increasing the ventilationi of the lungs are of preventiveand therapeutic value and should include frequent change of position, deepl)reathing exercises, and carbon dioxide-oxygen inlhalations. Carbon dioxide-oxygen inhalations are particularly helpful in increasing the ventilation of thelungs in those patients who are unable to obtain lhyperventilation by voluntarydeep breathing exercises.

The prompt recognition of the presence of bronchial secretions is offundamental importance in the prevention of postoperative pneumonia oratelectasis, for the reason that retained bronchial secretions are, in our opinion,a precursor to the developmnent of atelectasis and bronchopneumonia in mostinstances. The patient slhould be encouraged to cough at periodic intervalsof at least every two hours, not only to increase the ventilation of the lungs,but, of equal importance, to determine whetlher the cough is wet or dry. Awet type of cough signifies the presence of bronchial secretions, and demandsthat the secretions be evacuated, either by coughing, wlhich will usually beeffective, or by actual suction. The character of the respiratory sounds shouldbe elicited by auscultation with the stethoscope placed close to the patient'smouth, in order to determine wlhether the breath sounds are dry or whetherrlhonchi or wheezes are present. The presence or absence of rlhonchi shouldalso be determined by palpation and auscultation of the chest. As mentionedabove, auscultation of the bases of the lungs is preferably done with thepatient lying on alternate sides, both bases being examined while the patientis on each side, but particular attention being directed to the physical signsover the base of the uppermost lung. Due to the increased costal excursionof the uppermost lung and the resultant better ventilation of the lower lobeon this side, the breath sounds are heard to better advantage and rhonchi andwheezes are more often audible when the patient is in this position than whenhe is in the supine position. The clharacter of the breath sounds prior to,during and following a diagnostic cough should be elicited, as rhonchi thatare not present on deep breatlhing are often lheard during the inspiration andexpiration coincident witlh cough. WVhen patients are unable to ventilate the

250


bases well on deep breathing and when the cough is weak, the presence orabsence of rhonchi at the pulmonary bases should be determined by auscultationduring hyperventilation with carbon dioxide-oxygen inhalations. The earlydetection of rhonchi is of importance because their presence is the earliest signof partial bronchial obstruction due to secretions, occurring before the develop-ment of frank signs and symptoms of atelectasis or bronchopneumonia.

Voluntary cough is the most important single measure in the preventionand treatment of postoperative atelectasis and bronchopneumonia. The im-portance of an effective diagnostic or therapeutic cough should be stressed tothe patient, in order to obtain his cooperation, even though the act of coughingmay be attended with considerable discomfort. The patient should be in-structed to take several deep breaths before each cough in order to increaseits effectiveness. When the patient is in the supine position a painful abdom-inal incision should be supported by the nurse or surgeon by gentle constantpressure at each side of the incision, supplemented by firm compression of thecostal margins. The patient should also be shown how he can aid himselfby supporting the incision.

The position of the patient in bed is important in influencing the effective-ness and ease of cough. When he is in the supine position, the cough isusually more effective if the head of the bed is raised slightly than if the bedis flat or in the Trendelenburg position. Coughing is, however, usually easierand even more effective when the patient is in the lateral position. Manualsupport of the incision with the patient in this position is aided by the nurseor surgeon standing behind the patient, supporting the incision with onehand23 and exerting counterpressure over the spine with the other hand. Inthe lateral position the patient can conveniently support the incision with onehand, which in turn can be supported by the nurse's or surgeon's hand. Whentenacious secretions are being raised with difficulty, it is often necessary tooffer continuous verbal encouragement to the patient, as well as manual sup-port to the incision, so that the complete expulsion of secretions will beobtained. It is not sufficient merely to ask the patient to cough; he shouldbe instructed how to do so and helped to do so.

The use of a small dose of an opiate (insufficient to obtund the coughreflex) is of advantage in decreasing the amount of pain coincident withcoughing, thereby increasing the effectiveness of cough. Steam inhalationswith menthol aid in reducing the viscosity of bronchial secretions and are ofdefinite advantage when tenacious secretions are present. The continuous useof a steam tent provides another means for accomplishing the same purpose.Expectorants are of value and are used when they can be administered orally.When purulent bronchial secretions are present, the use of one of the sulfona-mide drugs is indicated; our preference at present is for sulfathiazole in viewof the fact that pneumococci are usually present.7 28

THE ASPIRATION OF BRONCHIAL SECRETIONS BY SUCTION

The prompt removal of bronchial secretions by suction is indicated assoon as it is evident that the cough is ineffectual and bronchial secretions are

251


beinig retainied. Two metlhods are available for the removal of bronclhialsecretions, one by bronchosco)y17 and the otlher by meanis of a catheter intro-duced into the tracheobronchial tree.'4 Bronchoscopy possesses the advantageof allowing visual inspection of the bronchi, and by its use onie can be certainthat the tracheobronchial tree is dry at the conclusion of the aspiration. Incases of postoperative atelectasis and bronchopneumonia, bronchoscopy re-veals varying amounts of thick, viscid secretion which at first is mucoid andlater mucopurulent. Often a tremendously large amount of secretion ispresent. The secretion is usually in a semifluid state and only partiallyoccludes the large bronchi; an actual mucous plug completely occluding abronchial orifice is an infrequent occurrence. Slight swelling and inflamma-tion of the tracheal and bronchial mucosa are usually seen on bronchoscopicexamination, and occasionally there are areas of moderate or a considerabledegree of edema of the mucosa of the lobar orifices. Bronchoscopy allowsthe inflamed mucosa to be shrunken with a solutioni of pontocaine and adrena-lin, and the lumen of the bronclhi is thereby enilarged. Bronchoscopy alsooccasionally reveals a localized adlherent fibrinous exudate which can beremoved by aspiration or by other mechanical means (Fig. 5). Our broncho-scopic findings are the same as those of otlhers in similar cases.4 3 16, 18, 21

Tracheobronchial suction by miieans of a catlheter introduced through thenose and into the bronchial tree provides a readily available measure for theaspiration of retained bronchial secretion.14 This mnethod, which has previ-ously been designated as intratracheal suction is more accurately describedas tracheobronchial suction, in that the large bronchi, as well as the trachea,are aspirated. Tracheobronchial suction can be used as an alternative tobronchoscopy in most instanices wlhen removal of secretions is indicated. Itis particularly applicable wlhen repeated aspirations are required, perhaps athourly or two-hourly intervals. Tracheobronchial suction provides a measurewlhich is usually quickly available when emergency aspiration of secretions isindicated, as it does not require the short delay necessary for the assemblingof bronchoscopic equipment. When the amount of retained bronchial secre-tion is small, tracheobronchial suction will usually be sufficient. When theamount of secretion is large, we prefer bronchoscopic aspiration as the initialprocedure, followed by catheter suction at frequent intervals until voluntarycough becomes effective. Occasionally, difficulty will be experienced in intro-(lucing the catheter into the trachea, and in suclh instances bronchoscopy shouldbe resorted to without delay. If a patient is critically ill and cyanotic, bron-choscopy, if expeditiously performed, is frequently a less upsetting measure,and oxygen can be conveniently admiinistered through the aspirating channelof the bronchoscope wlhile the tracheobronchial tree is being cleared by meansof the aspirator introdtuced througlh the bronchoscope. Oxygen, however, canlalso be given tlhrouglh a nasopharyngeal catheter durinig catheter suction ofthe bronchial tree.

The requirements for tracheobronchial suctioni are a No. i6 F. soft rubberuretlhral catlheter, a suction apparatus delivering I 5 to 25 lbs. suction, and

252


connecting tubing. We prefer a catheter of the Robinson type with twoopenings, and it should preferably be new and not softened by repeatedsterilizations. A Luken's glass bronchoscopic collecting tube is customarily

A. B.

0~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ R . .

C. D.FIG. 5.-L. H., NO. 425951: (A) Roentgenogram reveals pneumonitis and partial atelec-

tasis, right lung, 28 hours following cholecystectomy and choledochostomy. Large amountof purulent secretion aspirated by. tracheobronchial suction on two occasions. (B) Roentgeno-gram on following day. Clinical improvement but persistence of fever and large amount ofpurulent expectoration. In view of atelectasis of right middle lobe, bronchoscopy was believedpreferable to tracheobronchial suction, so that middle lobe orifice could be inspected during theaspiration. Bronchoscopy revealed partial obstruction of orifice of middle lobe bronchus byfibrinous exudate, which was removed. (C) Roentgenogram three days later demonstratesresidual infiltration of right middle lobe. Convalescence satisfactory. (D) Roentgenogramnine days later shows clearing of infiltration.

interposed in the system, in order to estimate the amount and consistency ofthe secretions and to obtain a specimen for bacteriologic examination. Theopen end of the collecting tube is fitted with a rubber stopper with a one-

253

HAIGHT AND RANSOM Annals of SurgeryAugust. 1941

quarter-inclh hole. The hole in the stopper is intermlittenitly occluded withthe thumb so that suctioni will be obtained only when the hole is occluded.If the collecting tube is not available, intermittent suction can be obtained bypinching and alternately releasing the connecting tubing, or by the use of aglass Y-tube, the open end of which is intermittently occluded with the finger.

..

A. B.FIG. 6.-(A) Apparatus bor ti acheobronchial sulction consists of SUCt;OnI machine, bronchoscopi)c

collecting tube (op)tional ) and No. I6 F. urethral catheter. (B) Bronchial secretionls, inure p)urulenltth. in usual, remuoed by tracheobronchial suction.

\AS the amounlt of sectretion obtained is freqluenltly mzoie than the bi onlchoscopic collecting tubJe can holdl, a pus-trap should be interposed into the systemto prevenlt the secretions from enterinlg the suiction apparatus (Fig. 6).

Thle catheter iS inltroduced inltO the tracheobronlchial tree without the useof local anesthesia. The patient is p)laced in the semi-Fowler position, theneck is flexedl slightly and the tongue is pulled forward by the operator inordler to elevate the epiglottis. The catheter is then introduced through thenose, USinlg the side whlich is the more widely patent, and it is directed pos-teriorly until the operator feels it touching the larynx. The catheter is thenwithdrawn I or 2 cm. (Fig. 7) and the patient is asked to take a quick deepbreath. The catheter is then quickly advanced into the trachea during deepinspiration. If this maneuver is unsuccessful, the patient is asked to coughand the catheter is quickly advanced during the deep inspiration followingcough. UnIless pharyngeal secretions are present, suction is not applied untilthe catheter has been introduced into the trachea. During the introductionof the catheter into the trachea, the operator maintains traction upon thetongue in order to prevent the patient from swallowing. The operator isassured that the catheter is in the trachea and not in the esophagus by the

254


onset of coughing, the passage of air through the catheter, or by huskiness ofthe voice when the patient is asked to speak.

FIG. 7.-Diagram illustrating method for introduction of catheter intotrachea. Tongue pulled forward to raise epiglottis, thereby opening passagewayfor catheter.

After the catheter is in thetrachea, the head of the bed is low-ered to the horizontal position.Suction is then applied for sev-eral seconds and tracheal secre-tions, if present, are aspirated.The suction is then stopped andthe patient is asked to take severaldeep breaths, following which thesuction is again applied for sev-eral seconds. This sequence isrepeated until the trachea is dry.The catheter is then introducedinto the bronchial tree of themore involved lung. Ordinarily,the catheter enters the rightbronchus, as it is more nearly inthe axis of the trachea. In orderto direct the catheter into the leftbronchus, the patient's chin andhead are turned far to the right

FIG. 8.-Roentgenogram showing chin and headturned to right, thereby directing catheter into leftbronchial tree.

255

HAIGHT AND RANSOM Annals of SurgeryHAIGHTANDRANSOM ~~August. 1941

(Fig. 8). Eaclh bronchial tree is aspirated dry, the aspiration being applied in-termittently to prevent excessive coughing and cyanlosis. The catlheter is intro-duced until it reaches the approximate level of the orifice of tlle lower lobebronchus, at wlhiclh timiie the outer end of the catheter will be al)out two tothree inches from the external nares. If one side shows a greater involvementthan the other, the patient may be rolled onto the contralateral side during

A. B.FIG. g.-S. B., No. 42207I: (A) Roentgenogram 36 hours following bilateral

inguinal herniorrhaphy. Atelectasis, right lung. Bronchoscopy performed same daybecause ot ineffectual cough and retained bronchial secretions. 25 cc. thick muco-purulent secretion aspirated from trachea and right bronchial tree. Cough remainedineffectual. Accordingly, two subsequent aspirations by tracheobronchial suction onsame day. Two aspirations by same method on following day until cough becameeffective. Convalescence satisfactory thereafter. (B) Roentgenogram six days later re-veals no pulmonary atelectasis or infiltration.

the aspiration of the more involved lung, so that posture will aid cough indislodging secretions from the smaller bronchi up to the larger bronchi, wherethey can be reached by the aspirating catheter. Ordinarily, the procedureof tracheobronchial suction is accompanied by moderate or considerable cough-ing, which is helpful in raising secretions from the smaller bronchi to thelevel where they can be removed. It should be emphasized that, unless inter-mittently applied, the suction will provoke severe coughing and cyanosis, and,accordingly, the suction should be used for periods of only several secondseach, the patient being allowed to take a few deep breaths between each periodof suction. The total duration of the procedure varies from two to threeminutes, depending upon the amount of secretions present. The amount ofsecretions aspirated is frequently larger than might be anticipated, the averagequantity usually being between 10 to 20 cc.

Following the use of bronchoscopy or tracheobronchial suction, voluntarycough usually becomes more effective due to the improved ventilation of thelung beyond the sites of the obstructing secretions. The improved aerationof the lung makes available a larger quantity of air to be displaced by thebechic blast, and peripheral secretions are raised with greater ease. Voluntary

256


cough, however, may not become completely effective until the patient's gen-eral condition improves.

As the subsequent formation of secretions is to be expected, the patientshould be carefully observed for any evidences of retained secretions, andsuction should again be instituted when and if there is further retention ofsecretions (Fig. 9). Owing to the improved bronchial drainage and to chemo-therapeutic measures, the accompanying purulent bronchitis subsides and thereformation of secretions gradually lessens. Also, during this interval, thepatient's general condition, unless influenced by extrapulmonary complications,gradually improves, and the voluntary cough becomes more effective. Themeasures mentioned earlier for aiding voluntary cough should be continued,so that voluntary cough will become effective as soon as possible and the needfor suction will not be unduly prolonged.

The improvement in the patient's condition following the removal ofretained secretions is often striking, especially when the amount of retainedsecretions may inadvertently have progressed to an alarming degree beforerecognition. Retained bronchial secretions are obstructing secretions, inter-fering with the airway to the lungs and preventing adequate ventilation.Accordingly, patients are able to breathe more comfortably following theaspiration of secretions, and cyanosis, if present before the aspiration, willfrequently be relieved by the improved pulmonary ventilation. As postopera-tive atelectasis and bronchopneumonia are usually sequelae of retained secre-tions, the prompt removal of secretions at the first evidence of their retentionwill minimize the incidence of these complications and lessen their severity.

SUMMARY

The presence of bronchial secretions, and the decreased pulmonary ventila-tion and cough efficiency subsequent to operation are vitally important factorsin the genesis of postoperative atelectasis and bronchopneumonia. The promptrecognition of retained bronchial secretions is essential in the prevention ofthese complications, and the methods for eliciting the physical signs ofretained secretions are described. The measures for aiding the cough mecha-nism are discussed and they are frequently effective in providing adequatedrainage of the tracheobronchial tree. As retention of bronchial secretionsoccurs in some instances in spite of these measures, retained secretions shouldbe immediately removed by bronchoscopy or tracheobronchial suction beforethe advanced signs of progressing bronchial obstruction and pulmonary infec-tion have developed. A technic for tracheobronchial suction is described; thesimplicity of the technic and its applicability when repeated aspirations arerequired, merit its more frequent use.

REFERENCESAlexander, J.: Preoperative and Postoperative Care of Patients with Surgical Diseasesof the Chest. Arch. Surg., 40, II33, I940.

2 Beecher, H. K.: The Measured Effect of Laparotomy on the Respiration. Jour. Clin.Invest., I2, 639, 1933.

257

HAIGHT AND RANSOM Annals of SurgeryHAIGHTANDRANSOM ~~August. 1941

3 Beecher, H. K., Bradshaw, H. H., and Lindskog, G.: Effect of Laparotomy andAbdominal Distention on Lung Volume. Jour. Thoracic Surg., 2, 444, I933.

4 Brown, A. L.: Postoperative Pulmonary Atelectasis: Observations on the Importanceof Different Types of Bronchial Secretion and Anesthesia. Arch. Surg., 22, 976,I93'.

5 Brunn, H., and Brill, S.: Observations on Postoperative Pulmonary Atelectasis:Consideration of Some-Factors in Its Etiology, Prevention and Treatment. ANNALSOF SURGERY, 92, 80I, 1930.

c Churchill, E. D., and McNeil, D.: The Reduction in Vital Capacity Following Opera-tion. Surg., Gynec., and Obstet., 44, 483, I927.

7 Coryllos, P. N.: Postoperative Pulmonary Complications and Bronchial Obstruction.Surg., Gynec., and Obstet., 50, 795, I930.

8 Cutler, E. C., and Hunt, A. M.: Postoperative Pulmonary Complications. Arch. Surg.,I, I14, 1920.

9 Cutler, E. C.: The Operative Efforts to Be Directed Toward the Prevention of Pul-monary Complications Through the Embolic Route. Internat. Abst. Surg., 68, 340,I939.

10 Elkin, D. C.: Postoperative Pulmonary Complications. Surg., Gynec., and Obstet.,70, 491, I940.

1 Faulkner, W. B., Jr.: Internal Drainage: Its Application in Pulmonary Suppuration.J.A.M.A., 95, I325, I930.

12 Fazekas, I. G.: Uber die Entstehung der postoperativen Pneumonie. Beitr. Z. klin.Chir., i69, I09, I939.

'3 Gray, H. K.: Postoperative Pulmonary Complications and the Postoperative Use ofthe Trendelenburg Position. Minnesota Med., i8, 273, 1935.

14 Haight, C.: Intratracheal Suction in the Management of Postoperative PulmonaryComplications. ANNALS OF SURGERY, 107, 2I8, I938; Surgery, 6, 445, 1939.

15 Hinshaw, H. C.: Methods for Reducing the Risk of Abdominal Operation in Casesof Pulmonary Disease. Surg. Clin. North America, 20, 973, I940.

16 Hollinger, P. H.: Bronchoscopy in Postoperative Pulmonary Complications. Surg.Clin. North America, I8, 237, I938.

17 Jackson, C., and Lee, W. E.: Acute Massive Collapse of the Lungs. ANNALS OFSURGERY, 82, 364, 1925.

18 Jackson, C., and Jackson, C. L.: Bronchoscopical Observations on Postoperative Pul-monary Complications. ANNALS OF SURGERY, 97, 5I6, 1933.

19Jones, D. F., and McClure, W. L.: The Influence of the Transverse Upper AbdominalIncision on the Incidence of Postoperative Pulmonary Complications. Surg., Gynec.,and Obstet., SI, 208, I930.

20 King, D. S.: Postoperative Pulmonary Complications. (i) A Statistical Study Basedon Two Years' Personal Observation. Surg., Gynec., and Obstet., 56, 43, I933.

21 Lee, W. E., Tucker, G., and Clerf, L.: Postoperative Pulmonary Atelectasis. ANNALSOF SURGERY, 88, 6, I928.

22 Lindskog, G.: Postoperative Pulmonary Complications: A Statistical Study Basedon Personal Observation of I,2I5 Consecutive Major Operations. Yale Jour. Biol.and Med., 9, 403, 1937.

23 Moore, A. E.: The Treatment of Postoperative Pulmonary Atelectasis. Surgery, 5,420, I939.

24 Muller, G. P., Overholt, R. H., and Pendergrass, E. P.: Postoperative PulmonaryHypoventilation. Arch. Surg., I9, I322, 1929.

25 Rovenstine, E. A., and Taylor, I. B.: Postoperative Respiratory Complications: Occur-rence Following 7,874 Anesthesias. Am. Jour. Med. Sci., igi, 807, I936;

26 Sante, L.: Massive (Atelectatic) Collapse of the Lung, with Especial Reference toTreatment. J.A.M.A., 88, I539, 1927.

27 Schmidt, E. R., and Waters, R M.: Anesthesia, Anesthetic Agents and Surgeons.Surgery, 6, 177, 1939.

258

Volume 114 ATELECTASIS AND BRONCHOPNEUMONIANumber 2

28 Whipple, A. 0.: A Study of Postoperative Pneumonitis. Surg., Gynec., and Obstet.,26, 29, I918.

DIscusSION.-DR. ELLIoTT C. CUTLER (Boston, Mass.): One of the topics set fordiscussion at the Stockholm meeting of the International Surgical Society this summerwas "Postoperative Pulmonary Complications." I take the contribution of DoctorsRansom and Haight as an evidence of the indestructibility of science at the hands evenof totalitarian warfare.

If we view postoperative pulmonary complications as a whole, we may find that atleast we know something about it, though, since my first studies with Doctor Morton,in I9I6, I feel there are still very large gaps. With the advent of local anesthesia, wehave evidence that it matters not what the anesthetic is, and the reports from clinicseverywhere in the world have come to this point of view. The same pulmonary compli-cations occur under local anesthesia to-day that occurred in i9oo under the hands ofMikulicz, who wrote the first paper on postoperative pulmonary complications underlocal anesthesia.

If we study the body as a whole, and take fields of surgery and instances of compli-cations in special fields, we learn a great deal. We find that the general percentage ofpulmonary complications for all surgery is between 2.5 and 3 per cent, but when wecome to the abdomen, it rises to I2 per cent; and when we come to the epigastrium itrises, in some clinics, to as high as 30 per cent. This is of considerable significance,because it must bear some relation, therefore, to the operative field and something thatgoes on in the operative field.

Having disabused ourselves, therefore, of the idea that the anesthesia plays a r6le,and having very definite evidence that, apparently, the complications bear a direct rela-tion to the ability of the patient to carry out normal respiration, we have left only oneother common factor, which is that these complications occur perhaps 5 to IO per centless frequently in females than in males. That has been interpreted by most workers asdue to the fact that a man breathes largely with his diaphragm and a woman is a costalrespiratory animal.

The proof of these contentions lies in part with adequate studies of vital capacity,and we finished the study of several thousand consecutive surgical operations on whomvital capacity studies have been done daily from the time of admission to the time ofdischarge, over a period of two and one-half years, and we have had adequate curvesof reduction in vital capacity with every surgical procedure.

The reduction varies from a median decrease around 59 to 6o per cent for the epi-gastrium to practically nothing for the extremities; this correlates, almost exactly, withthe incidence of pulmonary complications as a whole.

We felt it was wise to find some remedy for the painful respiration; and sought newdrugs and new methods of anesthesia. About the only one that has yielded any benefitis a combination of a long-lasting local anesthetic-eucupin with oil. Under these con-ditions, eucupin produces anesthesia often lasting for four or five days. In a limitednumber of patients having upper abdominal incisions, carefully studied by one of mycolleagues, we find that, when eucupin is used to block the field, the reduction in vitalcapacity is about one-half of that in similar patients, similarly operated upon by the samesurgeon, when eucupin is not employed to block the field.

It seems reasonable to suppose that if an adequate local anesthesia could be foundwhich would render epigastric wounds painless, and respiration would remain normal,we might greatly curtail this disastrous complication for the surgeon.

The authors of this paper have largely restricted their discussion, as we see, toatelectasis. Now, it is difficult to say what the relative frequency of the various clinicalforms of pulmonary complications are. If one includes large pulmonary emboli, thepulmonary emboli constitute about IO to I2 per cent of the complications. Lobar pneu-monia, proven by bacterial study, constitutes only 2 or 3 per cent; and you can dividethe rest between the clinical diagnosis of atelectasis, either massive or scattered, orbronchial pneumonia. But if you make a clinical diagnosis of scattered atelectasis, andif the patient comes to autopsy, the professor of pathology may tell you it is bronchialpneumonia. Whether that is only because the late stages of atelectasis reach consolida-tion or not, we do not know.

The chief gift of this paper to-day-and everything has its gift to science-seemsto me to be this ingenious idea of using such a simple method in the wet, blue patient,

259

HAIGHT AND RANSOM Annals of Sur1er1as the installationi of a catheter into the trachea; and adequate instructions on how todo this have been given. Most of us, I am sure, suck out the mouth, but the additionof tracheal suctionI will do much.

I have only one other suggestioni, which is that if the reduction in vital capacityfrom pain is an important objective, it is peculiar that all the hibernatinig animals, fromthe bear to the bat, do not have this disease, because their respiratory rate is cut downland the vital capacity is almost nothinig. So I return again to an old contelntion of mine-that these complications may result from the promulgation of small emboli up thepassageways of the lymphatic current through the pleura of the diaphragm.

DR. WALTER ESTELL LEE (Philadelphia, Pa.): I hesitate to discuss this signlificantpaper because of my too frequent contributions to the subject in the past.

However, such a paper seems timely, for in our experience, and I suspect it is thesame with others, the incidenice of postoperative atelectasis is increasinig. This, of course,may be due, in part at least, to a more general recognitioni of the conditioni by bothphysicians and surgeons, but it seems to us, in most part, to be the result of more radicaland more proloniged surgical procedures.

We nlow speak of six and oine-half- and seveni-hour operations, such as Doctor Laheyreported recenitly, and to mainltain anesthesia and complete muscular relaxationi for suchperiods of time, spinal, and particularly conltinluous (iintermittenlt) aniesthesia is beitngemployed almost routinely in many clinics. With such a method of anesthesia, mostsurgeons are usinig more and more sedatives in the form of morphine, or some type ofbarbiturate. Under these conditions the respiratory movements are far more shallowthan in inhalation anesthesia, and the cough reflex is depressed or abolished. Undersuch conditions it is to be expected that the tracheal and the bronchial secretions willtend to accumulate in the dependetnt portions of the bronchial tree and even in the alveoli.

One should approach the problem of postoperative atelectasis through prophylaxisrather than by treatment, and though the method which Doctor Haight has outlined isideal, we would suggest that prophylaxis should start with the aniesthetist during theoperation, wheni every effort should be made to maintaini, at all times, an unobstructedbronchial airway, and not wait until the close of the operation to start aspiration. Thiscan be donie very readily by hypervenitilation with oxygeln under pressure about every I5minutes, and if there is any excess of secretion, it should be aspirated during this tinme,and not wait until the close of the operative procedure.

My appeal is that we should not postpone Doctor Haiglht's suggestion of trachealdrainage until the signls of bronchial obstructioni appear postoperatively, but that routiniemeasures in the form of hyperventilation, with oxygen under pressure, durinlg the opera-tion, and that aspiration of excessive amounts of tracheal secretion at the close of theoperation and before the patient leaves the operating table should be practiced routinely.

In our original reports we confessed-probably bragged-that we have performedbronchoscopic drainage in some 8o patients during a period of one year. At the presenttime, in a much larger group of patients, three to five bronchoscopic drainages a yearis our average.

Bronchiti_s _ __ Chei cal DR. FRANK H. LAHEY (Boston,Infectiouis Mass.): I uldertake further discus-

sioIn only because I think this is such4,o \ an importallt subject and because itMassive collapse has p)layed such1 a part in reducing ourAt1ecants ---Lobar collapseAtelectasis - -Lobular collapse mortality.Lobular collapse Albert M4iller, of Providence, has

showin that if you subj ect people to

4,f / anesthesia of any type, which is suf-/Aec i pficient in depth to produce relaxation,"Aeettc ~they breatlhe largely witlh the dia-

)phragln. Tracings shlow that mostBronchopneusonia respiration is carried oni by the clia-

CIIART I. phragnm, wvhether spinal, nitrous oxideor ether anesthesia is used. So, as Doctor Lee has stated, a great deal of our difficulty isprobably due to this.

260

Volume 114Nutmber 2 ATELECTASIS AND BRONCHOPNEUMONIA

I thinlk we owe a trelmenidouis amiounlt to Clhevalier Jacksoni an(l Gabriel rucker,because, so long ago, they called our attenitionl to the value of suictioln bronchoscopy. Onehas onily to see the intrathoracic goiter patienits, (leliriouis in the midd(le of the nlight, thesecretion sucked out ancd the teml)erature dowli the next (lay ali(l the patieiit in a rationalstate, to realize what an important part this plays in the prevelltion of pulmonary compli-cations.

Chart i is a diagrammatic scheme of the productioni of these bronchial pneumoniasand atelectases. Here is a repetition of xvhat we all have seen, atelectasis so graphicallycleared by suction bronchoscopy. Chart 2 shows the temperature and pulse reactions.We could repeat this time after time. There are certain warniings that I think onie shouldnote. Our anesthetists should be trained in suction bronchoscopy and, in turn, shouldtrain our Fellows in catheter bronchoscopy.

Another point which I think impor-tant is that bronchoscopy should beundertaken in the middle of the night,when the condition is discovered, andnot at a convenient time the next day.We have demonstrated, in our autopsyfindinigs, how rapidly pneumonitis candevelop, and I believe that we shouldhave available the medical men, theroentgenologists, and the suction bron-choscopists to do it, not when it isconvenient the next day but wheneverthe evidence occurs. These patientswho have been delirious in the middleof the night, who have had the secre-tions sucked out and have become con-scious and rational, when they havedifficulty with their breathing andmucus have even requested its repeti-tion themselves because they have beenso much improved. One only has tosee the striking results that come fromcatheter and particularly suction bron-choscopy to be impressed with the factthat this is a real contribution and will,I believe, save many lives. It hasplayed a very important part, I am cer-tain, in making it possible for us tomaintain such a low mortality ratein the subtotal gastrectomy cases.

I..a,2 3 41 1-61Co718.9

t - -1- -- - e-- - -

97_< T{

,j go s _--f-----4 .- --fF

32O1 -^s II-CHART 2.

DR. HENRY K. RANSOM (Ann Arbor, Mich., closing): I wish to mention brieflytwo points: First, that conservative measures are of inestimable value in the preventionof postoperative pulmonary complications due to retained bronchial secretions. Suchmeasures include (i) the avoidance of excessive sedation and prolonged periods of semi-consciousness following operation; (2) frequent changes in the positioIl of the patient;(3) encouraging the patient to cough, usiIng carbon dioxide inhalations if necessary; (4)good nursing care; and (5) a cooperative house staff. By attention to these details theprocess can often be aborted and thus prevented from going on to the more advancedand serious stages. These same measures will also be sufficient in many of the earlycases of established atelectasis to afford proper drainage of the bronchial tree, and therebyto effect a cure, and, therefore, the more complicated procedures described by DoctorHaight will only occasionally be necessary. Second, during the past several years,transverse incisions have become increasinigly popular ill the field of abdominal surgerv,and especially for operations upon the biliary tract. For the past three years we haveemployed such incisions and the results have been gratifying. It seems to be true thatafter healing has taken place, the abdominal wall is more sound, that the incidence ofincisional hernia is decreased, that patients are more comfortable during the postoperativeperiod, and that they may be allowed out of bed somewhat earlier. We also gained the

261


impression that the:e was a lower incidence of pulmonary complications where suchincisions were used. Some ten years ago, a similar observation was recorded in an articleby Jones and McClure. They stated that in the series of cases reported by them no in-stances of postoperative pneumonia or atelectasis occurred. We, therefore, reviewed ourcases of simple cholecystectomy during the past six years in order to make a comparisonof the pulmonary complications noted with vertical and with transverse incisions (Table I).

TABLE I

PULMONARY COMPLICATIONS FOLLOWING CHOLECYSTECTOMY FOR

CHRONIC CHOLECYSTITIS, WITH OR WITHOUT CHOLELITHIASIS1935-I94I

Type of IncisionVertical Transverse

Number of cases ............. ................................. 346 io8Patchy atelectasis or pneumonia .................... 26 (7.5%) 5 (4.6%)Massive atelectasis . . 5 oPleuritis ......................................... I OInfarct or embolism ........ ....................... II

33 (9.3%) 6 (s.6%)Pulmonary complication chief cause of death......... 3 0Pulmonary complication contributory cause of death.. I 0

These statistics show that while there was a slight decrease in the incidence ofpulmonary complications when transverse incisions were used, this decrease was not asgreat as we had anticipated. Since the number of cases is relatively small, the statisticalevidence is not of great significance, but it does suggest that this is one more detail illsurgical technic which may be of importance in further reducing the number of thesedreaded postoperative complications.

262

Documents

POSTOPERATIVE ATELECTASIS AND BRONCHOPNEUMONIA